Review of Benedict Carey’s “How We Learn: The Surprising Truth About When, Where, and Why It Happens”.

When you read, does testing help material stick in your head? Suppose you and a friend read a poem for a few minutes, and then shut the book, never re-reading it. One of you tries reciting it from memory soon after, the other doesn’t. A week later, would one of you remember more of the poem?

Suppose you want to perfect three different tennis serves. You have enough time for practicing 3,000 serves. Would you gain more from “blocked practice” — repeating the same serve hundreds of times consecutively — or from mixed practice — cycling through the various types of serves?

The book describes over a century of fascinating experiments, with researchers grappling with what aids learning, what sticks, and how well that learning translates to related endeavors. Chapters talk about the role of context (summary: things learned in a range of settings and moods is retained longer), about spacing in study (five sessions of ten minutes spaced over days is better than one long session of comparable duration), about self testing (testing is not just for evaluation: it actively helps learning), about marination of ideas (down time and relaxation helps on certain tasks, and different kinds of tasks respond better to active relaxation — surfing — vs. passive relaxation — napping), and about the positive role of interruption (incomplete tasks have a share of our attention, and one semester-long project can act to focus the mind better than multiple smaller projects that, having been completed early in the semester, recede from attention). At the same time, this book is a fascinating study of development in this science, which, like all other sciences, is full of dead ends, of esoteric squabbles grinding progress to a halt, of decade long quixotic quests, of poignant tales of promising researchers quagmired in Stalinist Russia, stuff that makes for gripping narrative.

In the tennis example above, a curious paradox emerges. Here and in learning more academic material and in mastering more esoteric pursuits as recognizing an artist’s style — identifying a painting as a Goya or a Monet or an O’Keeffe — blocked study is demonstrably less effective than mixing, and yet the students’ self reports of what worked suggest that they thought that they were faster progressing in the blocked case.

Perhaps the blocked regime provides a false sense of progress and lulls the mind into thinking that a skill has been learned. Not knowing the shortcomings in what we have learned, the mind does not bother fixing what ain’t broken. Both self testing and mixed practice reveals gaps, and allows us to fill them.

The book has been effective in one way: it woke me from a slumber of passive consumption of a range of ideas, making a strong case for how the act of describing something you have read can show gaps in understanding, or perhaps the nonexistence of coherent understanding, and explains why I started writing today. I am leery of saying that this awakening is for good and I will write everyday or whatever — that claim had better wait for a few weeks of regular writing under my belt.

I do unsupervised concept discovery at Pinterest (and previously at Google). Twitter: @amahabal

I do unsupervised concept discovery at Pinterest (and previously at Google). Twitter: @amahabal